Dataset Information

Scalable Database Indexing and Fast Image Retrieval Based on Deep Learning and Hierarchically Nested Structure Applied to Remote Sensing and Plant Biology.

ABSTRACT: Digitalisation has opened a wealth of new data opportunities by revolutionizing how images are captured. Although the cost of data generation is no longer a major concern, the data management and processing have become a bottleneck. Any successful visual trait system requires automated data structuring and a data retrieval model to manage, search, and retrieve unstructured and complex image data. This paper investigates a highly scalable and computationally efficient image retrieval system for real-time content-based searching through large-scale image repositories in the domain of remote sensing and plant biology. Images are processed independently without considering any relevant context between sub-sets of images. We utilize a deep Convolutional Neural Network (CNN) model as a feature extractor to derive deep feature representations from the imaging data. In addition, we propose an effective scheme to optimize data structure that can facilitate faster querying at search time based on the hierarchically nested structure and recursive similarity measurements. A thorough series of tests were carried out for plant identification and high-resolution remote sensing data to evaluate the accuracy and the computational efficiency of the proposed approach against other content-based image retrieval (CBIR) techniques, such as the bag of visual words (BOVW) and multiple feature fusion techniques. The results demonstrate that the proposed scheme is effective and considerably faster than conventional indexing structures.

SUBMITTER: Sadeghi-Tehran P

PROVIDER: S-EPMC8320911 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Computational pan-genomics utilizes information from multiple individual genomes in large-scale comparative analysis. Genetic variation between case-controls, ethnic groups, or species can be discovered thoroughly using pan-genomes of such subpopulations. Whole-genome sequencing (WGS) data volumes are growing rapidly, making genomic data compression and indexing methods very important. Despite current space-efficient repetitive sequence compression and indexing methods, the deployed compression methods are often sequential, computationally time-consuming, and do not provide efficient sequence alignment performance on vast collections of genomes such as pan-genomes. For performing rapid analytics with the ever-growing genomics data, data compression and indexing methods have to exploit distributed and parallel computing more efficiently. Instead of strict genome data compression methods, we will focus on the efficient construction of a compressed index for pan-genomes. Compressed hybrid-index enables fast sequence alignments to several genomes at once while shrinking the index size significantly compared to traditional indexes. We propose a scalable distributed compressed hybrid-indexing method for large genomic data sets enabling pan-genome-based sequence search and read alignment capabilities. We show the scalability of our tool, DHPGIndex, by executing experiments in a distributed Apache Spark-based computing cluster comprising 448 cores distributed over 26 nodes. The experiments have been performed both with human and bacterial genomes. DHPGIndex built a BLAST index for n = 250 human pan-genome with an 870:1 compression ratio (CR) in 342 minutes and a Bowtie2 index with 157:1 CR in 397 minutes. For n = 1,000 human pan-genome, the BLAST index was built in 1520 minutes with 532:1 CR and the Bowtie2 index in 1938 minutes with 76:1 CR. Bowtie2 aligned 14.6 GB of paired-end reads to the compressed (n = 1,000) index in 31.7 minutes on a single node. Compressing n = 13,375,031 (488 GB) GenBank database to BLAST index resulted in CR of 62:1 in 575 minutes. BLASTing 189,864 Crispr-Cas9 gRNA target sequences (23 MB in total) to the compressed index of human pan-genome (n = 1,000) finished in 45 minutes on a single node. 30 MB mixed bacterial sequences were (n = 599) were blasted to the compressed index of 488 GB GenBank database (n = 13,375,031) in 26 minutes on 25 nodes. 78 MB mixed sequences (n = 4,167) were blasted to the compressed index of 18 GB E. coli sequence database (n = 745,409) in 5.4 minutes on a single node.

Dataset Information

Scalable Database Indexing and Fast Image Retrieval Based on Deep Learning and Hierarchically Nested Structure Applied to Remote Sensing and Plant Biology.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets